A workshop is proposed to be held in Tarrytown, New York, in the spring of 1995, two years after the previous workshop. Its topic is Statistical Methods in Genetic Mapping. Twenty-seven participants will be invited (including the two principal investigators) to outline specific statistical aspects and problems relevant to gene mapping. The following are examples of problems to be addressed: Optimal design strategies for genome-wide mapping of disease genes (two-stage and multistage mapping); marker errors, their multipoint linkage analysis between a disease locus and highly polymorphic markers; integration of new CA repeat maps with established restriction fragment length polymorphism maps. The workshop is expected to result in new ideas and perhaps a consensus on new strategies for tackling some of the problems addressed. A workshop report will be submitted for publication in Genomics.